INTERPRETING RESULTS OF A GIS-BASED LANDSLIDE SUSCEPTIBILITY ASSESSMENT USING GEOTECHNICAL INVESTIGATIONS FOR A ROAD SECTOR IN NORTH-WESTERN ROMANIA

References: 14th International Multidisciplinary Scientific GeoConference SGEM 2014, www.sgem.org, SGEM2014 Conference Proceedings, ISBN 978-619-7105-12-4 / ISSN 1314-2704, June 19-25, 2014, Book 2, Vol. 3, 691-698 pp

ABSTRACT
A landslide susceptibility assessment was performed using the statistical method of logistic regression in a 100 km2 area from north-western Romania. The methodology was based on GIS techniques of spatial analysis, previously mapped landslides in the field and the assumption that future landslides will occur under the same circumstances as the ones from the past. The results were validated using an independent set of landslides and showed a good agreement with observations from the field. A project involving stabilising measures for a road sector crossing the study area and connecting urban centres of regional importance gave the opportunity of comparing the statistical results with geotechnical investigations performed at six drilling points by a local geotechnical firm. Further field observations led to a more detailed mapping of the two major landslides intersected by the road sector and the mapping of more recent, smaller landslides. The interpreting of the geotechnical investigation results concluded in identifying the relative depth of landslide slip surface at the analysed points, confirmed by the Bishop model at a relative depth of 4 and 3 m, and highlighted the main lithologic characteristics of the sedimentary deposits involved in landslide activity. Appropriate measures for slope stabilisation near the road sector were also produced in the project and critical considerations about the results of the statistical model used in the first stage of analysis are made. In addition, several interpretation restrictions are advised to local and regional users of the landslide susceptibility map.

Keywords: landslide susceptibility, logistic regression, GIS, geotechnical investigations